Metabolism of Inositol 1,4,5-Trisphosphate and Inositol 1,3,4,5-Tetrakisphosphate by the Oocytes of Xenopus laevis*

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 273, No. 7, Issue of February 13, pp , by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Metabolism of Inositol 1,4,5-Trisphosphate and Inositol 1,3,4,5-Tetrakisphosphate by the Oocytes of Xenopus laevis* (Received for publication, August 6, 1997, and in revised form, October 30, 1997) Christopher E. Sims and Nancy L. Allbritton From the Department of Physiology and Biophysics, University of California, Irvine, California The pathway and kinetics of inositol 1,4,5-trisphosphate (IP 3 ) metabolism were measured in Xenopus laevis oocytes and cytoplasmic extracts of oocytes. Degradation of microinjected IP 3 in intact oocytes was similar to that in the extracts containing comparable concentrations of IP 3 ([IP 3 ]). The rate and route of metabolism of IP 3 depended on the [IP 3 ] and the intracellular free Ca 2 concentration ([Ca 2 ]). At low [IP 3 ] (100 nm) and high [Ca 2 ](>1 M), IP 3 was metabolized predominantly by inositol 1,4,5-trisphosphate 3-kinase (3-kinase) with a half-life of 60 s. As the [IP 3 ] was increased, inositol polyphosphate 5-phosphatase (5-phosphatase) degraded progressively more IP 3.Ata[IP 3 ]of8 M or greater, the dephosphorylation of IP 3 was the dominant mode of IP 3 removal irrespective of the [Ca 2 ]. At low [IP 3 ] and low [Ca 2 ] (both <400 nm), the activities of the 5-phosphatase and 3-kinase were comparable. The calculated range of action of IP 3 in the oocyte was 300 m suggesting that IP 3 acts as a global messenger in oocytes. In contrast to IP 3, inositol 1,3,4,5-tetrakisphosphate (IP 4 ) was metabolized very slowly. The half-life of IP 4 (100 nm) was 30 min and independent of the [Ca 2 ]. IP 4 may act to sustain Ca 2 signals initiated by IP 3. The halflife of both IP 3 and IP 4 in Xenopus oocytes was an order of magnitude or greater than that in small mammalian cells. Ca 2 signals regulate a diverse array of cellular functions including secretion, cytoskeletal rearrangement, and gene transcription (1, 2). Modulations in [Ca 2 ] 1 are used to transduce signals in nearly all cells including those of plants and animals (1 3). Generation and degradation of the second messenger IP 3, which opens the IP 3 receptor/ca 2 channel on the endoplasmic reticulum, regulates the formation and termination of Ca 2 signals. Knowing the rates and pathways of IP 3 degradation is fundamental to understanding Ca 2 wave formation. The metabolism of IP 3 and its products by phosphatases and kinases is unfolding as an increasingly complex process * This work was supported by the Arnold and Mabel Beckman Foundation, the Searle Scholars Program, and National Institute of Mental Health Grant MH The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. To whom correspondence should be addressed. Tel.: ; Fax: The abbreviations used are: [Ca 2 ], free Ca 2 concentration; I-1-P 1, inositol 1-phosphate; IP 1, all isomers of inositol monophosphate; I-1,4- P 2, inositol 1,4-bisphosphate; IP 2, all isomers of inositol bisphosphate; IP 3, inositol 1,4,5-trisphosphate; I-1,3,4-P 3, inositol 1,3,4-trisphosphate; IP 4, inositol 1,3,4,5-tetrakisphosphate; [IP 3 ], concentration of IP 3 ; [IP 4 ], concentration of IP 4 ;IP 2, inositol, IP 1, and IP 2 ; [IP 2 ], the sum of the concentration of inositol, IP 1 and IP 2 ; 5-phosphatase, inositol polyphosphate 5-phosphatase; 3-kinase, inositol 1,4,5-trisphosphate 3-kinase; HPLC, high pressure liquid chromatography; BAPTA, 1,2-bis(2-aminophenoxy)ethane-N,N,N,N -tetraacetic acid (4 6). Two primary degradative pathways exist for IP 3, but they differ in their relative importance among cell types and between species. Several isoforms of the 5-phosphatase dephosphorylate IP 3 yielding inositol 1,4-bisphosphate (4 8). IP 3 is also a substrate of the 3-kinase which phosphorylates IP 3 to form IP 4 (9 13). The binding of Ca 2 /calmodulin to the 3-kinase enhances its activity (14 16). The 5-phosphatase also metabolizes IP 4 to inositol 1,3,4-trisphosphate (I-1,3,4-P 3 ). Current evidence suggests that the major function of the 5-phosphatase in the phosphoinositide cycle is to decrease the [IP 3 ] and the concentration of IP 4 ([IP 4 ]). In contrast, the role of the 3-kinase is to generate another second messenger, IP 4,as well as to decrease the [IP 3 ]. An increasing amount of evidence suggests that IP 4 is an important regulatory molecule in cells (6). Like IP 3, IP 4 may be involved in the regulation of the [Ca 2 ] (17 19). IP 4 binds to the IP 3 receptor/ca 2 channel and releases Ca 2 from the endoplasmic reticulum although with a 10-fold lower potency than IP 3. More intriguing, IP 4 binds with high affinity to several intracellular proteins, synaptotagmin I and II, Gap1, Btk, and centaurin- (20 24). The Ras GTPaseactivity of Gap1 is stimulated by IP 4, and IP 4 may interact with synaptotagmin to inhibit synaptic transmission (20, 25). The pathway selected to metabolize IP 3 may not only influence its rate of removal but also alter subsequent signal transduction within the cell. Much of our mechanistic knowledge of Ca 2 wave propagation is derived from studies of Ca 2 waves in oocytes and eggs of Xenopus laevis. Formation of the Ca 2 wave, which follows the fertilization of an egg and which follows the activation of plasma-membrane receptors in oocytes, requires IP 3 (26, 27). Work to date on IP 3 metabolism in Xenopus oocytes has yielded conflicting results. Microinjection of concentrated [ 3 H]IP 3 into single Xenopus oocytes followed by separation of the metabolites by ion-exchange chromatography has suggested that the 5-phosphatase pathway prevailed (28). However, the addition of [ 3 H]IP 3 to a homogenate of oocytes or microinjection into ovarian follicles followed also by high pressure liquid chromatography (HPLC) separation have provided strong evidence for the 3-kinase as the primary route of IP 3 degradation (29, 30). These discrepancies remain unresolved, despite the central role that the Xenopus oocyte has played in understanding Ca 2 wave propagation. All three of the studies discussed above found a surprisingly prolonged degradation time for IP 3,20 min or longer. This contrasts with smaller cells that degrade IP 3 in a few seconds (31, 32). Establishing both the rate and pathway of IP 3 metabolism in oocytes is required to understand the role of the phosphoinositide pathway in the generation and termination of Ca 2 signals and in interactions with other signal transduction cascades. The purpose of this investigation was to determine how IP 3 and IP 4 were degraded in X. laevis oocytes. Moreover, the goal was to quantitate the kinetic properties of these metabolic pathways in the cytoplasmic milieu. This paper is available on line at

2 Metabolism of IP 3 and IP 4 by X. laevis Oocytes 4053 EXPERIMENTAL PROCEDURES Materials Tritiated inositol and inositol phosphates were purchased from NEN Life Science Products. Nonradioactive IP 3 was purchased from Calbiochem and Alexis (Woburn, MA), and nonradioactive IP 4 and calpain inhibitor I (N-Ac-Leu-Leu-norleucinal) were purchased from Calbiochem. Phorbol 12-myristate 13-acetate was supplied by Alexis (Woburn, MA). EGTA puriss grade, was obtained from Fluka (Ronkonkoma, NY). Rhod-2, BAPTA, and calcium green-5n were supplied by Molecular Probes (Eugene, OR). All other reagents were purchased from Fisher. EGTA-buffered Ca 2 Solutions An equimolar solution of EGTA and Ca 2 was prepared by the method of Tsien and Pozzan (33). Cytoplasmic extracts with varying [Ca 2 ] were made by addition of 10 mm EGTA and 10 mm EGTA with 10 mm Ca 2. The [Ca 2 ] in the EGTA-buffered extracts was estimated from the [Ca 2 ] in buffer A (135 mm KCl, 5 mm NaCl, 1 mm MgCl 2,10mM HEPES, ph 7.4) which approximated the intracellular ionic environment and contained the same mixture of 10 mm EGTA and 10 mm EGTA with 10 mm Ca 2 as the EGTA-buffered extract. The [Ca 2 ] in the buffer solution was measured using the fluorescent Ca 2 indicators, rhod-2 and calcium green-5n as described by Allbritton and colleagues (34) and Haugland (35). The [Ca 2 ]in buffer A containing 10 mm EGTA and 10 mm Ca 2 was 10 M. Preparation of Oocytes and Cytoplasmic Extracts Female X. laevis frogs were purchased from Nasco (Modesto, CA). Oocytes were surgically obtained and prepared as described previously (34, 36). Cytoplasmic extracts were also made as described previously with the following exceptions (34, 37). To minimize proteolysis the oocytes and extract were maintained at 4 C throughout the preparation. Calpain inhibitor I (10 g/ml) was added to the extract to block the Ca 2 -activated protease calpain. After isolation, the cytoplasmic extract was used immediately in experiments. Prolonged delays prior to use diminished the ability of the extract to metabolize inositol phosphates. This cytoplasmic preparation is 90% pure cytoplasm (90 mg/ml protein) (34). Measurement of IP 3 Degradation in Intact Oocytes Oocytes were microinjected with 5 30 nl of [ 3 H]IP 3 (2 M) contained in buffer A with or without BAPTA (100 mm). The volume of injectate was determined from the total number of counts contained in the oocyte. The oocytes were incubated at room temperature for the indicated times. Just prior ( 10 s) to the end of the incubation period, the buffer solution surrounding the oocyte was removed to determine how much of the tritium label had leaked from the oocyte. Cells that leaked greater than 10% of the radioactive label were excluded from experiments. The intracellular reactions of the oocyte were terminated by rapid freezing with powdered dry ice. The frozen oocyte was then homogenized in chloroform: methanol (50 l of 1:2) and buffer B (50 l of25mm tetrabutylammonium hydrogen sulfate, 20 mm KH 2 PO 4, ph 3.5) with 80 g of hydrolyzed phytic acid added as a carrier to prevent nonspecific loss of inositol phosphates. Hydrolyzed phytic acid was prepared as described by Irvine and colleagues (38). After addition of chloroform (50 l), the mixture was centrifuged at 15,000 g, and the aqueous phase was separated by reverse phase HPLC. Measurement of Inositol Phosphate Degradation in Cytoplasmic Extracts [Ca 2 ] in the cytosol was buffered at the indicated concentration by addition of 10 mm EGTA and 10 mm EGTA with 10 mm Ca 2.In some instances, phorbol 12-myristate 13-acetate (50 ng/ml) was included in the cytoplasmic mixture. [ 3 H]IP 3 or [ 3 H]IP 4 was then added to the oocyte cytosol. After the addition of each reagent to the cytosolic extract, it was gently mixed over a 5-s period by pipetting the mixture up and down. The cytoplasmic extract was incubated for the indicated time at room temperature ( 20 C). The reaction was stopped with a 16-fold excess volume of trifluoroacetic acid (10%) containing hydrolyzed phytic acid (0.8 mg/ml) and centrifuged at 15,000 g. The supernatant was dried, resuspended in buffer B, and separated by reverse phase HPLC. Reverse Phase HPLC Inositol phosphates were separated on an analytical C18 column (Alltech, San Jose, CA) maintained at 45 C with an elution protocol modified from that of Shayman and Bement (39) and Sulpice et al. (40). The flow rate was 1 ml/min. During the first 22 min, fractions were collected every 0.25 min and thereafter every 1.25 min. To monitor the separation efficiency of the column, a set of standards previously added to and then extracted from cytosol was chromatographed each day. [ 3 H]Inositol coeluted with inositol 1-phosphate as did the other isomers of inositol phosphate. The isomers of inositol bisphosphate were not distinguishable. RESULTS To qualitatively determine the rate of IP 3 metabolism in vivo, stage V/VI oocytes were microinjected with [ 3 H]IP 3 (2 M) (n 11). After incubation for 30 s, 1 min, or 5 min, the oocytes were rapidly frozen to terminate intracellular reactions. The inositol phosphates were extracted from the cytoplasm and separated by HPLC. Representative standard and experimental traces are shown in Fig. 1, A C. Oocytes microinjected with small volumes of [ 3 H]IP 3 metabolized half of the [ 3 H]IP 3 in approximately 1 min (Fig. 1B). Longer times were required to degrade half of the [ 3 H]IP 3 when larger volumes were microinjected (Fig. 1C). In all of the experiments, the metabolic fraction that contained the most tritium was that of IP 4 or IP 3. In vivo and at an [IP 3 ]of2 M and lower, IP 3 was metabolized predominantly by the 3-kinase. In other species the activity of the 3-kinase is increased by the binding of calmodulin and Ca 2 (11, 14 16). To determine whether [Ca 2 ] regulated the metabolism of IP 3 in Xenopus oocytes, [ 3 H]IP 3 (2 M) was coinjected with BAPTA (100 mm) (n 10) to diminish the IP 3 -mediated increase in [Ca 2 ]. After a 1-, 5-, 15-, or 30-min incubation time, the inositol phosphates were extracted and separated. Remarkably, very little of the [ 3 H]IP 3 was metabolized by 5 min (Fig. 1D). The half-life of the microinjected IP 3 was 13 min. In contrast to the oocytes without BAPTA, substantial amounts of tritium did not accumulate in the IP 4 fraction (Fig. 1, D and E). The activity of the 3-kinase was dramatically decreased in the oocytes containing BAPTA presumably due to a decrease in [Ca 2 ]. In these in vivo experiments, both diffusion and degradation altered [IP 3 ] during the time course of the measurements. One minute after microinjection in the absence of degradation, the [IP 3 ]atthe microinjection site would still be 10 times greater than the final equilibrium concentration (41). Since [IP 3 ] was also altered by diffusion, these experiments could not be used to measure quantitatively the rates and pathways of IP 3 metabolism. To provide a quantitative description of the metabolism of IP 3, subsequent experiments were performed in a cytosolic extract. In past experiments, cytosolic preparations were greatly diluted during preparation, and the compartmentalization of proteins and membranes was abolished by homogenization (29, 30). To eliminate these disadvantages, we used a cytosolic extract that is 90% of the concentration of undiluted cytoplasm (34). The cytosol contains nearly all of the intracellular organelles, and they retain many of their normal functions including the ability to transit through the cell cycle (34, 37, 42). These observations suggested that this extract could be used to define how [IP 3 ] and [Ca 2 ] regulate the removal of IP 3 in an environment similar to the cytoplasmic milieu. [ 3 H]IP 3 (100 nm) was added to a cytoplasmic extract that contained 10 mm EGTA with 10 mm Ca 2 ([Ca 2 ] 10 M). At varying times an aliquot of this cytosolic mixture was removed, and the inositol phosphates were extracted and chromatographed. The half-life of the [ 3 H]IP 3 was 60 s (Fig. 2B). Under these conditions most of the [ 3 H]IP 3 was converted to IP 4. The half-life of IP 4 was considerably longer than that of IP 3. For incubation times of 5 min or less, the metabolism of IP 4 was negligible (see also Fig. 4). By 10 min much of the tritium ( 30 40%) initially added to the cytosol was no longer soluble after acid extraction; the [ 3 H]inositol was recycled into a cellular component other than inositol or an inositol phosphate. These results are consistent with those obtained in the intact oocytes microinjected with [ 3 H]IP 3 suggesting that the cytosolic extract is a very good model for the intact oocyte. For experimental time points of 5 min or less, the cytosolic reactions of IP 3 could be segregated into two distinct pathways without common metabolites (Fig. 2A). In the 3-kinase path-

3 4054 Metabolism of IP 3 and IP 4 by X. laevis Oocytes FIG. 1.Metabolism of IP 3 by intact oocytes. An oocyte was microinjected with [ 3 H]IP 3 (7 nl (B) or25nl(c)) or [ 3 H]IP 3 and BAPTA (14 nl (D) or 20 nl (E)). After 1 (B),5(C and D), or 15 (E) min, the oocyte was frozen rapidly, and the inositol phosphates were extracted from the oocyte and chromatographed. Representative HPLC traces are shown in B E. An HPLC trace of tritiated standards ( 50 nci each) which were added to and then extracted from cytoplasm is shown in A. The standards were inositol 1-phosphate (I-1-P 1 ) (fraction 11), inositol 1,4-bisphosphate (I-1,4-P 2 ) (fraction 19), I-1,3,4-P 3 (fraction 68), IP 3 (fraction 74), IP 4 (fraction 92), and IP 6 (fraction 105). B E, the migration times of the accompanying set of standards are marked by arrows. The migration times of the standards in A C differ from those in D and E since the samples were chromatographed on different HPLC systems. The standards eluted in the same order in all experiments. Since IP 6 was not included in the standard mix for the experiments in D and E, the migration time of IP 6 was not marked in these panels. way, IP 4 accumulated with very little conversion of IP 4 to the lower inositol phosphates. Consequently, [IP 4 ] was used as an estimate of the activity of the 3-kinase for time points of 5 min or less. Inositol bisphosphate (IP 2 ) was then formed almost exclusively by dephosphorylation of IP 3 rather than by dephosphorylation of I-1,3,4-P 3. Inositol and inositol monophosphate

4 Metabolism of IP 3 and IP 4 by X. laevis Oocytes 4055 (IP 1 ) were formed by subsequent dephosphorylation of that IP 2. The products of the 5-phosphatase pathway were inositol, IP 1, and IP 2, and the sum of the concentrations of these three inositols ([IP 2 ]) was used to estimate the activity of the 5-phosphatase toward IP 3. For times less than 5 min, neither inositol, IP 1, nor IP 2 accrued to a substantial degree (Fig. 2B). Neither of these inositol fractions exceeded 10% of the total tritium in the cytosol. The metabolite of IP 4, I-1,3,4-P 3, did not accumulate either. The chromatographic peak representing this isomer never contained more than 6% of the added tritium. Although both the 3-kinase and 5-phosphatase were active in the extract, the conversion of IP 3 to IP 4 dominated at these low [IP 3 ] and for incubation times less than 5 min. Since the activity of the 5-phosphatase was much lower than that of the 3-kinase, the kinetic properties of the 3-kinase were estimated from the rate of formation of IP 4. The kinetics of this reaction appeared first order; the rate of formation of IP 4 was proportional to [IP 3 ] (43). The first order rate constant (k) obtained by fitting the [IP 4 ] versus time trace to an exponential was 10 2 s 1. Since the prior in vivo experiments suggested that [Ca 2 ] regulated the metabolism of IP 3 in oocytes, [ 3 H]IP 3 (100 nm) was added to a cytoplasmic extract which contained 10 mm EGTA decreasing [Ca 2 ] to less than 100 nm (Fig. 2C). Since the half-life of IP 4 remained much greater than 5 min (see also Fig. 4), the data were displayed identically to that in Fig. 2B.At low [Ca 2 ] the half-life of IP 3 was increased markedly to 10 min. In agreement with the in vivo results, the longer half-life was due to a decrease in the activity of the 3-kinase. Less than 7 and 25% of the IP 3 was converted to IP 4 in 1 and 10 min, respectively. As before, I-1,3,4-P 3 did not accumulate. The rate of accrual of IP 4 and IP 2 was similar and linear over time (0.04 nm/s) and nearly identical to the rate of accumulation of IP 2 in the presence of high [Ca 2 ] (Fig. 2, B and C). Formation of IP 2 was independent of [Ca 2 ] suggesting that the 5-phosphatase of Xenopus oocytes was not regulated by [Ca 2 ]. The unchanged production of [IP 2 ], despite a markedly decreased [IP 4 ], also suggests that the lower inositol phosphates resulted predominantly from sequential dephosphorylation of IP 3 rather than IP 4. To determine the range of [Ca 2 ] that potentiated the 3-kinase s activity, we varied the [Ca 2 ] in the extract by altering the ratio of 10 mm EGTA to 10 mm EGTA with 10 mm Ca 2. After addition of [ 3 H]IP 3 (100 nm) to the cytosolic preparation for 1 min, the inositol phosphates were extracted and chromatographed. As the [Ca 2 ] was increased from 100 nm to 2 M, the amount of IP 3 degraded increased from 7 to 40 nm. Half-maximal activation of the 3-kinase occurred at a [Ca 2 ]of 390 nm and maximal activation appeared by 1 M (Fig. 2D). Several previous reports indicated that the 5-phosphatase was activated after phosphorylation by protein kinase C, whereas other experiments suggested that the 5-phosphatase activity was inactivated by protein kinase C (44 48). Addition of phorbol 12-myristate 13-acetate (50 ng/ml), a potent activator of protein kinase C, to cytosolic extracts with varying [Ca 2 ] did not alter the rate or pathway of IP 3 metabolism (data not FIG. 2.Metabolism of IP 3 by cytoplasmic extracts of oocytes. A, the schematic of IP 3 metabolism was valid for incubation times of 5 min and less when [Ca 2 ] in the extract was elevated ( 5 M) and for 10 min and less when [Ca 2 ] was very low ( 100 nm). B and C,[ 3 H]IP 3 (100 nm) was added to a cytoplasmic extract with [Ca 2 ] buffered to 10 M (B) or to less than 100 nm (C). After varying incubation times, the inositol phosphates were isolated and separated by HPLC. [IP x ] represents the concentration of IP 3 ( ), IP 4 (E), or IP 2 ( ). IP 2 represents the sum of the concentrations of inositol, inositol phosphate, and inositol bisphosphate. The traces are the best fits to the data points obtained with Origin (Microcal, Northampton, MA). B, the IP 4 and IP 3 data points were fit to an exponential function, whereas the IP 2 data points in B and all points in C were fit to a straight line. D, [ 3 H]IP 3 (100 nm) was added to a cytoplasmic extract with varying [Ca 2 ] and incubated for 1 min. The concentrations of the inositol phosphates were quantitated and graphed as in B and C. The traces are hand drawn.

5 4056 Metabolism of IP 3 and IP 4 by X. laevis Oocytes FIG. 4.Metabolism of IP 4 by cytoplasmic extracts of oocytes. [ 3 H]IP 4 (10 M) was added to a cytoplasmic extract which was then incubated for varying times. The inositol phosphates were isolated and quantitated. The concentrations of IP 2 (Œ, ), IP 4 (, E), and I-1,3,4-P 3 (f, ) are plotted against the incubation time. [Ca 2 ] in the extract was buffered to 10 M (, E, ) or to less than 100 nm (Œ,, f). The lines through the IP 4 and IP 2 data points are fits to an exponential function, and the line through the I-1,3,4-P 3 data points is hand drawn. FIG. 3.The effect of [IP 3 ] on the pathway of IP 3 metabolism. Varying concentrations of [ 3 H]IP 3 were added to a cytoplasmic extract with [Ca 2 ] buffered to 10 M (A) or to less than 100 nm (B). After 5 min the cytosolic reactions were stopped, and the tritiated inositol phosphates were isolated and quantitated. [IP 4 ] (E) and [IP 2 ] ( ) are plotted against the initial [IP 3 ] in the cytoplasmic extract. The traces through the data points are hand drawn. shown). Under these conditions, activation of protein kinase C did not alter the dephosphorylation of IP 3 by the 5-phosphatase or phosphorylation of IP 3 by the 3-kinase. The Michaelis constant (K m ) of the 5-phosphatase for IP 3 is greater than 10 times that of the 3-kinase in many different tissues (4, 5). Therefore, the metabolic pathway of IP 3 may depend on [IP 3 ]. To determine how the route of degradation in Xenopus oocytes changed with [IP 3 ], we varied the starting [IP 3 ] in the cytosolic preparation from 100 nm to 30 M. [ 3 H]IP 3 was incubated in the extract for 5 min followed by extraction and separation of the inositol phosphates. In the first series of experiments, [Ca 2 ] was maintained at 10 M by addition of 10 mm EGTA with 10 mm Ca 2. For a starting [IP 3 ] of less than 8 M, the major metabolic pathway was conversion to IP 4 by the 3-kinase (Fig. 3A). At an initial [IP 3 ] of approximately 2 M and greater, the rate of formation of IP 4 was independent of [IP 3 ] consistent with [IP 3 ] being much greater than the K m of the 3-kinase. The rate of generation of IP 4 at this high [IP 3 ] was used to estimate the maximal velocity (V max ) of the 3-kinase for IP 3 (4 nm/s). For a starting [IP 3 ] greater than 8 M, [IP 2 ] continued to increase despite a plateau in the generation of IP 4 (Fig. 3A). The additional IP 2 must originate from dephosphorylation of IP 3 by the 5-phosphatase. At high [IP 3 ], most of the IP 2 was formed by dephosphorylation of IP 3 rather than IP 4 ; IP 3 was degraded chiefly by the 5-phosphatase. To determine how a decrease in [Ca 2 ] altered IP 3 metabolism at different [IP 3 ], the initial [IP 3 ] was varied from 100 nm to 30 M in a cytosolic preparation containing 10 mm EGTA (Fig. 3B). The [ 3 H]IP 3 was incubated in the extract for 5 min followed by acid extraction and chromatographic separation. For all starting [IP 3 ], the rate of formation of IP 4 was greatly decreased compared with that in the presence of micromolar [Ca 2 ]. In contrast, the rate of accrual of [IP 2 ] was nearly unchanged from that in the presence of high [Ca 2 ]. As before the 5-phosphatase was independent of [Ca 2 ]. Moreover, [IP 2 ] did not depend on [IP 4 ] indicating that these lower inositol phosphates resulted mainly from the action of the 5-phosphatase on IP 3. The rate formation of IP 4 became independent of [IP 3 ] when the initial [IP 3 ] was 10 M or greater. This rate was used to estimate the V max of the 3-kinase at low [Ca 2 ](1 nm/s). For most initial [IP 3 ], IP 3 was degraded mainly by dephosphorylation, and the rate of formation of [IP 2 ] was linearly related to the initial [IP 3 ] (Fig. 3B). For an initial [IP 3 ]of 1 M or greater, [IP 3 ] decreased negligibly during the course of the reaction ( 20%). Since [IP 3 ] was nearly constant during the reaction time and [IP 2 ] was linearly related to the starting [IP 3 ], the reaction of the 5-phosphatase with IP 3 was modeled as a first order reaction for [IP 3 ] greater than 1 M. The k of the 5-phosphatase for IP 3 ( s 1 ) was then estimated from the rate of formation of [IP 2 ]. When [Ca 2 ] was in the micromolar range, the metabolic pathway of IP 3 depended on [IP 3 ]. At low initial [IP 3 ], IP 3 was metabolized predominantly by the 3-kinase, whereas at high initial [IP 3 ], the 5-phosphatase dominated. The crossover point from the 3-kinase to the 5-phosphatase occurred at an [IP 3 ] of approximately 8 M. Since IP 4 may be an important signaling molecule, the rate and pathway of IP 4 degradation were determined. [ 3 H]IP 4 (100 nm or 10 M) was added to a cytosolic preparation containing 10 mm EGTA with 10 mm Ca 2. At the indicated times the inositol phosphates were extracted from an aliquot of the reaction mixture and separated. For both initial [IP 4 ], 50% of the [ 3 H]IP 4 was degraded by dephosphorylation in 30 min (Fig. 4). After a 5-min reaction time, only 5% of the IP 4 was metabolized. For reaction times of 5 min or less, destruction of IP 4 was negligible, and the reaction scheme depicted in Fig. 2A for IP 3 was valid. Varying the [Ca 2 ] in the extract from 100 nm to 10

6 Metabolism of IP 3 and IP 4 by X. laevis Oocytes 4057 M did not alter the rate of degradation of IP 4. As in prior experiments, I-1,3,4-P 3 did not accumulate to a substantial degree although [IP 2 ] increased with time. Conversion of IP 4 to IP 3 was not observed. Metabolism of IP 4 and formation of IP 2 followed first order kinetics for initial IP 4 concentrations of 100 nm and 10 M. The k obtained by fitting the [IP 4 ]or [IP 2 ] curves to an exponential was s 1. DISCUSSION Cell type TABLE I The range of action of IP 3 Radius Half-life of IP 3 Range of action M s M Microvilli of olfactory neurons (52) Smooth muscle (53) Parotid acinar cells (32) Neuroblastoma cells (31) Skeletal muscle (53) 5 to Xenopus Oocytes to Intact oocytes and a cytoplasmic extract of oocytes were used to determine the pathways and estimate the rates of IP 3 and IP 4 metabolism in Xenopus oocytes. The pathway of IP 3 degradation depended on both [IP 3 ] and [Ca 2 ]. The 3-kinase was regulated by [Ca 2 ]; the measured V max and k of the 3-kinase decreased as [Ca 2 ] decreased. Lowering [Ca 2 ] from 1 M to less than 100 nm diminished the k and the V max by a factor of 25 and 4, respectively. The k of the 3-kinase for IP 3 in the presence of 10 mm EGTA ([Ca 2 ] 100 nm) was estimated from the slope of the IP 4 trace in Fig. 2C and the initial slope of the IP 4 trace in Fig. 3B (k s 1 ). The K m calculated from the V max and k increased from 400 nm to 2.5 M as [Ca 2 ] declined from greater than 1 M to less than 100 nm. [Ca 2 ] regulated the 3-kinase by altering both the V max and K m of the enzyme for IP 3. By releasing Ca 2,IP 3 enhances its own degradation. The K m values of the 3-kinase and its regulation by Ca 2 in Xenopus oocytes are close to that reported in other species (4, 5, 10 12). These similar enzymatic features suggest that the Xenopus oocyte 3-kinase is regulated by calmodulin as is the case in other species. The k value of the 5-phosphatase for IP 3 was approximately twice that of the 3-kinase when [Ca 2 ] was low ( 100 nm). However, when [Ca 2 ] was greater than 1 M, the k value of the 3-kinase exceeded that of the 5-phosphatase by almost 20-fold. Although the V max of the 5-phosphatase for IP 3 could not be determined, the V max must be greater than 16 nm s 1, the fastest rate measured for the 5-phosphatase (Fig. 3). The K m of the 5-phosphatase for IP 3 must then be greater than 27 M. The first order behavior of the 5-phosphatase at IP 3 concentrations up to 30 M also suggests that the K m is greater than 30 M. The lower limit of the K m value and the absence of regulation by [Ca 2 ] are consistent with the properties of the 5-phosphatase in other species (4 8, 48 50). For all [Ca 2 ], the K m and V max values of the 5-phosphatase for IP 3 were much greater than those of the 3-kinase for IP 3. The relative values of the K m and V max of the 5-phosphatase and 3-kinase divided IP 3 metabolism into three regions. The regions were defined by their [Ca 2 ] and [IP 3 ] and were characterized by their different metabolic routes for IP 3. At high [Ca 2 ]( 400 nm) but low [IP 3 ] ( 8 M), IP 3 was predominantly metabolized by the 3-kinase to IP 4. The 5-phosphatase and the 3-kinase degraded roughly similar amounts of IP 3 when both [Ca 2 ] and [IP 3 ] were low ( 400 nm and 1 M, respectively). However, the 5-phosphatase was the dominant metabolic enzyme when [IP 3 ] was greater than 8 M irrespective of [Ca 2 ]. The regions that occur physiologically are not known since the intracellular range of [IP 3 ] has yet to be defined. IP 4 was metabolized very slowly with a half-life of approximately 30 min. The k value of the 5-phosphatase for IP 4 was similar to that for IP 3. The reported K m of the 5-phosphatase for IP 4 in other species is approximately 1 M (7, 51). When the initial [IP 4 ] was 10 M, the metabolism of IP 4 followed first order kinetics. Therefore, it is unlikely that the K m of the 5-phosphatase for IP 4 in Xenopus oocytes was as low as 1 M. When [Ca 2 ] was elevated, phosphorylation of IP 3 to IP 4 was much faster than degradation of IP 4. Consequently, repeated or persistent stimulation of IP 3 production in the presence of high nanomolar to micromolar [Ca 2 ], could result in IP 4 concentrations that are far greater than that of IP 3. Although opening of the IP 3 receptor/ca 2 channel requires 10-fold more IP 4 than IP 3, the accumulation of IP 4 may result in physiologically important IP 4 -mediated Ca 2 release. IP 4 may play a necessary role in sustained signal transduction. The k and V max values of the oocyte s 3-kinase and 5-phosphatase were markedly lower than the values reported in other species (4 12, 48 50). Since IP 3 microinjected into intact oocytes was also metabolized slowly, the long IP 3 degradation times were not an artifact of the cytosolic preparation. Additionally, IP 4 microinjected into intact oocytes has a half-life similar to that reported here in the cytosolic extract (29, 30). The lower k and V max values of the Xenopus oocyte compared with those of mammalian cells may be a reflection of the different size, temperature, and physiologic role of these cells. Slowing the degradation of IP 3 may be important for signal transduction from the cell s surface to its interior. The 500- m radius of the oocyte makes the transmission of information from the plasma membrane to the central core formidable. Slowing the metabolism of IP 3 increases its range of action, the distance over which a spatially localized impulse of IP 3 can diffuse (34). The range of action (R) ofip 3 can be estimated from its diffusion coefficient (D) and half-life (t1 2) in cytoplasm: r (2 * D * t1 2) 1/2. D for IP 3 is 280 m 2 s 1, and the t1 2 for IP 3 in the oocyte ranges from 1 to greater than 5 min. R for IP 3 is 300 m suggesting that IP 3 acts as a long-ranged messenger in the oocyte. In nearly all cell types, IP 3 is a global messenger (Table I). Whereas the rate of degradation of IP 3 and the size of a cell vary over several orders of magnitude, the range of action of IP 3 is nearly the same as the cellular dimensions. IP 3 transmits signals received at the plasma membrane throughout the cell, coordinating cell-wide activities. Acknowledgments We gratefully acknowledge Tobias Meyer and Lubert Stryer for stimulating discussions and H. N. Dao for technical assistance. REFERENCES 1. Berridge, M. J. (1993) Nature 361, Clapham, D. E. (1995) Cell 80, Allen, G. J., Muir, S. R., and Sanders, D. (1995) Science 268, Majerus, P. W. (1992) Annu. Rev. Biochem. 61, Shears, S. B. (1992) Adv. Second Messenger Phosphoprotein Res. 26, Irvine, R. F. (1995) Biochem. Soc. Trans. 23, Mitchell, C. A., Connolly, T. M., and Majerus, P. W. (1989) J. Biol. Chem. 264, Hansen, C. A., Johanson, R. A., Williamson, M. T., and Williamson, J. R. (1987) J. Biol. Chem. 262, Irvine, R. F., Letcher, A. J., Heslop, J. P., and Berridge, M. J. (1986) Nature 320, Biden, T. J., and Wollheim, C. B. (1986) J. Biol. Chem. 261, Ryu, S. H., Lee, S. Y., Lee, K. Y., and Rhee, S. G. (1987) FASEB J. 1, Johanson, R. A., Hansen, C. A., and Williamson, J. R. (1988) J. Biol. Chem. 263, Daniel, J. L., Dangelmaier, C. A., and Smith, J. B. (1988) Biochem. J. 253, Biden, T. J., Peter-Riesch, B., Schlegel, W., and Wollheim, C. B. (1987) J. Biol. Chem. 262, Yamaguchi, K., Hirata, M., and Kuriyama, H. (1987) Biochem. J. 244, Morris, A. J., Downes, C. P., Hardin, T. K., and Mitchell, R. H. (1987) Biochem. J. 248, Loomis-Husselbee, J. W., Cullen, P. J., Dreikausen, U. E., Irvine, R. F., and Dawson, A. P. (1996) Biochem. J. 314, Parker, I., and Ivorra, I. (1991) J. Physiol. (Lond.) 433,

7 4058 Metabolism of IP 3 and IP 4 by X. laevis Oocytes 19. Ferguson, J. E., Han, J. K., Kao, J. P. Y., and Nuccitelli, R. (1991) Exp. Cell Res. 192, Cullen, P. J., Hsuan, J. J., Truong, O., Letcher, A. J., Jackson, T. R., Dawson, A. P., and Irvine, R. F. (1995) Nature 376, Fukuda, M., Aruga, J., Niinobe, M., Aimoto, S., and Mikoshiba, K. (1994) J. Biol. Chem. 269, Salim, K., Bottomley, M. J., Querfurth, E., Zvelebil, M. J., Gout, I., Scaife, R., Margolis, R. L., Gigg, R., Smith, C. I., Driscoll, P. C., Waterfield, M. D., and Panayotou, G. (1996) EMBO J. 15, Fukuda, M., Kojima, T., Kabayama, H., and Mikoshiba, K. (1996) J. Biol. Chem. 271, Hammonds-Odie, L. P., Jackson, T. R., Profit, A. A., Blader, I. J., Turck, C. W., Prestwich, G. D., and Theibert, A. B. (1996) J. Biol. Chem. 271, Llinas, R., Sugimori, M., Lang, E. J., Morita, M., Fukuda, M., Niinobe, M., and Mikoshiba, K. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, Nuccitelli, R., Yim, D. L., and Smart, T. (1993) Dev. Biol. 158, Lechleiter, J., Girard, S., Peralta, E., and Clapham, D. (1991) Science 252, Shapira, H., Lupu-Meiri, M., and Oron, Y. (1992) J. Basic Clin. Physiol. Pharmacol. 3, McIntosh, R. P., and McIntosh, J. E. A. (1990) Biochem. J. 268, McIntosh, R. P., and McIntosh, J. E. A. (1990) Biochem. Biophys. Res. Commun. 166, Wang, S. S., Alousi, A. A., and Thompson, S. H. (1995) J. Gen. Physiol. 105, Burgess, G. M., McKinney, J. S., Irvine, R. F., and Putney, J. W. (1985) Biochem. J. 232, Tsien, R., and Pozzan, T. (1989) Methods Enzymol. 172, Allbritton, N. L., Meyer, T., and Stryer, L. (1992) Science 258, Haugland, R. (1996) Handbook of Fluorescent Probes and Research Chemicals, (Spence, M. T. Z., ed) pp , 6th Ed., Molecular Probes, Inc., Eugene, OR 36. Zagotta, W. N., Hoshi, T., and Aldrich, R. W. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, Murray, A. W., and Kirschner, M. W. (1989) Nature 339, Wreggett, K. A., Howe, L. R., Moore, J. P., and Irvine, R. F. (1987) Biochem. J. 245, Shayman, J. A., and Bement, D. M. (1988) Biochem. Biophys. Res. Commun. 151, Sulpice, J. C., Bachelot, C., Gascard, P., and Giraud, F. (1990) Methods in Inositide Research, (Irvine, R. F., ed) pp , Raven Press, New York 41. Crank, J. (1989) The Mathematics of Diffusion, p. 29, Clarendon Press, Oxford 42. Newmeyer, D. D., and Wilson, K. L. (1991) Methods Cell Biol. 36, Segel, I. H. (1975) Biochemical Calculations, pp , 2nd Ed., John Wiley and Sons, New York 44. Biden, T. J., Valler, L., and Wollheim, C. B. (1988) Biochem. J. 251, King, W. G., and Rittenhouse, S. E. (1989) J. Biol. Chem. 264, Connolly, T. M., Lawing, W. J., and Majerus, P. W. (1986) Cell 46, Lin, A. N., Barnes, S., and Wallace, R. W. (1990) Biochem. Biophys. Res. Commun. 170, Sim, S. S., Kim, J. W., and Rhee, S. G. (1990) J. Biol. Chem. 265, Connolly, T. M., Bross, T. E., and Majerus, P. W. (1985) J. Biol. Chem. 260, Downes, C. P., Mussat, M. C., and Michell, R. H. (1982) Biochem. J. 203, Connolly, T. M., Bansal, V. S., Bross, T. E., Irvine, R. F, and Majerus, P. W. (1987) J. Biol. Chem. 262, Breer, H., Boekhoff, I., and Tareilus, E. (1990) Nature 345, Walker, J. W., Somlyo, A. V., Goldman, Y. E., Somlyo, A. P., and Trentham, D. R. (1987) Nature 327,